AI.info

Zhibin Duan

Explore Zhibin Duan on AI.info.

• Mitigating Reward Hacking in RLHF via Bayesian Non-negative Reward Modeling